We Are Undervaluing James Paxton

Meandering through the first batches of ADP data is like Christmas Morning.

Some gifts you were expecting and are thankful to have gotten. We can compare these to Mike Trout and Kris Bryant. You knew they would be first round values and your suspicions are confirmed, content with the sanity of owners across the nation.

Other gifts you were not expecting, and for good reason. These are you classic pair of socks, the gift you leave under the tree to gather dust. Seeing value elsewhere, you are a little wary about spending an early pick on these guys. This list will vary by owner, but for myself in 2017, I feel myself being passive on Jonathan Villar and Buster Posey.

Then you stumble on the good stuff. The gifts you remember mentioning once to a family member, but never thought they would deliver with such thoughtfulness.

This year, that gift is James Paxton.

My gift to you? A quick blind player comparison.

As you could have guessed, James Paxton is one of these players.

In revealing the other two, we will take our first look at some ADP data.

Comparison statistics can draw a lot of criticism because of their selectiveness, and tendency not to tell the whole story about specific players. My intention of using them in this case is to illustrate how good of a season that Paxton put together, and on a higher level, start your wheels on wondering where his ADP should be heading into draft season.

Diving In...

Statistics suggest that a full prior season of data is more predictive than either half of a given season. The crux I always have with this finding are the specific cases in which mechanics for a pitcher or hitter change, resulting in noticeably different performance compared to the player’s prior results. A good example of this on the offensive side is Daniel Murphy.

Our initial claim would suggest that Murphy’s 2015 full season stats with the Mets were a better predictor for 2016 than either sample of first or second half data. Yet some analysts out there noticed a mechanical change with Murphy in his historic 2015 postseason run and pondered that a power spike could be on the menu in 2016 (the most recent in my memory was in this episode of Flags Fly Forever).

The toughest part is determining when a substantial enough mechanical change occurs, and how much to favor the more recent peripheral data over the prior results with the old mechanics.

James Paxton is an animal who falls squarely into this realm. A switch to his natural arm slot, apparent to even the most unskilled eye, resulted in a substantial improvement from June onward in 2016. Eno Sarris does a great job of explaining the changes, which took place after 50 innings in the minors to start the year. There are numerous graphs to cite that show Paxton’s change, so below are two simple ones, along with a comparison between Paxton’s results before and after this change (mechanical change implemented June 1st, to initially discouraging results).

Our first graph (thanks to BrooksBaseball.net) shows you the decrease in vertical height of Paxton’s release, which as you move to the second graph, correlates well with the change in velocity. Jumps are most noticeable from the prior year in Paxton’s fastball and cutter/slider, two pitches that he combined to throw nearly 78% of the time.

The most important thing in these results is the halving of Paxton’s walk rate, an indication that his natural arm slot provides much better command. Eno goes drastically increased ability to spot his fastball inside and cutter/slider on the outside part of the plate with increased consistency. A recipe for success if Paxton wants to excel in 2017.

Even with these improvements, there are still some concerns.

Paxton gets hit at a decent rate, evidenced by the fact that his walk rate was below 5% in 2016, but his WHIP lingered around 1.31. The signs pointing to an improvement in the high WHIP for 2017 is the sky high .347 BABIP Paxton carried on his belt last season. This is due in part to the -27.9 DEF rating the Mariners as a team accrued from their below average defensive performance on the season. BABIP is not guaranteed to fall in the following year, a common misconception among fantasy owners, but the fact that Paxton only had slightly above average line drive and hard hit rates, makes me confident in at least some decrease.

Relative to the rest of the league, Paxton does stand out. Among pitchers with the lowest 25 FIP totals in 2016 and 100 innings of work or more, Paxton also possessed the second lowest strand rate at 66.3%, about seven percentage points lower than the league average. This could be due in part to the high average against he relinquished, but could also relate to an inconsistency with his new arm slot from the stretch.

Time to finally acknowledge the elephant in the room, Paxton’s health.

This is by far the most volatile aspect of any fantasy asset’s stock. Starting Pitchers are getting injured at a scary rate relative to the past few years, and in comparison to position players, which makes an investment in one at the top end of drafts all the more risky. Investing in upside with pitchers like Paxton becomes even more appealing.

Despite the tag of ‘injury-prone’ that comes in tandem with Paxton’s name, he managed to throw 171 innings between his minor league stint and the majors. Paxton still managed to hit the DL in August, but it was from an elbow contusion he developed via an unavoidable line drive. Add the jump in fastball velocity, which is unfortunately tied to both success and injury, and we have a lefty with a smattering of concerns, but none so dire that it breeds confidence in disaster.

While we can choose to sit on the precedent that injuries are simply more common with Paxton, I am inclined to think a more comfortable arm slot has to help the health of his arm, or at least mitigate the stress that an unnatural arm slot would place on Paxton.

Fantasy Value

Now that we’ve trudged through 900+ words and five graphs of information, we can circle back to the matter at hand.

Where should Paxton actually be drafted?

In the initial stages of my site’s fantasy baseball rankings, BigThreeSports.com, we have moved Paxton up above his current NFBC ADP, more than any other player. The qualifier being that we have not finished our rankings to set in stone Paxton as our ‘Holy Grail’ of value.

Steamer currently has Paxton pegged for 171 innings with a 3.43 ERA and around 165 strikeouts. It also envisions improvement in both Paxton’s BABIP and strand rate, which pushes his projected WHIP down about one tenth of a point, a nice improvement.

11W/165K/3.43ERA/1.20WHIP is a likely outcome. But I like upside, and feel projections for Paxton in 2017 are a decent base, but fail to fully encapsulate the impact this mechanical change can have. We can guess the line above is about the mean outcome, but what if we kick up our expectation?

14W/180K/3.30ERA/1.15WHIP in around 185-190 innings, the workload being the least believable out of these metrics. We can call this the 70th percentile of Paxton’s fantasy output in 2017.

Kenta Maeda posted a 16W/179K/3.48ERA/1.14WHIP, which made him a top 20 pitcher in 2016 according to ESPN’s roto player rater. The 20th SP is being drafted between the 7th and 8th rounds on average, with Maeda himself going about one round later than that at 105 overall.

Should Paxton go ahead of Maeda? Probably not. Maeda has a full season of success and a very nice floor based on his command and repertoire of pitches. Paxton on the other hand, has the risk we have touched upon and a lot to prove. But too much of that risk is being built into his draft stock at 207 overall.

Right now, I have Paxton as my 32nd starting pitcher off the board (, about 70-80 spots ahead of his current ADP in roto leagues.

I am comfortable with taking him near this mid-11th round value. It may seem insane, but some guys I have no issue in putting Paxton over are Alex Reyes, Michael Fulmer, Kevin Gasuman, and J.A. Happ.

Here are two of our discussed Paxton outcomes compared to what Steamer has for the other four pitchers, all going in that 150 overall range.

We have risk with Paxton, no doubt, but is it substantial enough after a 170 inning season to eclipse other risky arms like Rich Hill or Matt Harvey?

How about the four pitchers above? Reyes comes into this season with sky high dynasty value, and my anointment as the number two fantasy prospect in baseball, but his control is an issue that I think will be a serious problem in 2017. Fulmer I might have faith in with points leagues, but his lack of strikeouts does not make me confident in his roto value. Gasuman is probably the pitcher of these four I am most excited about, yes the success of the Oriole youngster isn’t as good as what Paxton has shown. J.A. Happ’s stock is driven so heavily by wins and contact that he falls into the Fulmer arena of points league targets. Give me Paxton over all of these.

Investing in risk later in draft is less of a burden. If you can read your draft room right and Paxton falls past 170th overall, you have struck gold. Otherwise I do not think it is a reach to grab our lefty in the 11th round.